Do changes in obesity affect total testosterone levels? longitudinal results from the Massachusetts male aging study

Purpose Obesity has become a major public health challenge in the United States, now exceeding even smoking as a contributor to disease processes. Obesity has been inversely correlated with total testosterone levels. Do changes in obesity level affect total testosterone levels? Methods Using data from the Massachusetts Male Aging Study, a prospective, community-based, random sample study of men aged 40 to 70 at baseline, we investigated whether changes in obesity affect total testosterone levels. At baseline (T1; 1987–1989) and a subsequent follow-up (T3; 2002–2004), serum total testosterone was measured on morning blood samples. Obesity was defined as body mass index BMI ≥ 30 kg/m2. Four obesity change categories were constructed: obese at neither time point, at T1 only, at T3 only, and at both T1 and T3. Of the 1709 men recruited at T1, 855 participated at T3. Analysis of variance was used to model T3 testosterone level as a function of obesity change with adjustment for T1 testosterone level, age, chronic illness, medication, alcohol, smoking, and physical activity. Pairwise comparisons between the Neither group and the other obesity change categories were made using the Bonferroni procedure to adjust for multiple comparisons. Results The number of men in the Neither, T1 only, T3 only, and T1 and T3 groups was 409 (69%), 17 (3%), 84 (14%), and 80 (14%), respectively. Change in obesity was significantly associated with T3 testosterone level in both crude (P < 0.0001) and adjusted (P < 0.0001) models. Adjusted means for T3 testosterone level were: Neither, 422 ng/dL (95% CI = 408–436); T1 only, 422 ng/dL (95% CI = 353–491), T3 only, 348 ng/dL (95% CI = 317–379); and T1 and T3, 373 ng/dL (95% CI = 341–405). Both the T3 only (P < 0.0002) and the T1 and T3 (P < 0.04) group means were significantly lower than the Neither group mean. Conclusion Men who became obese or stayed obese had significantly lower total testosterone levels than men who were not obese at T1 or T3. Reducing obesity may be an effective (and safer) alternative to testosterone replacement therapy for improving total testosterone levels of aging men.